Metabolic Alterations in Blood and Saliva During the First Two Cycles of Adjuvant Chemotherapy Correlate with 5-FU Dose Reduction Requirement in Patients with Colorectal Cancer.

BACKGROUND AND OBJECTIVE: Personalized therapy strategies based on patient characteristics and informative biomarkers may reduce the risk of chemotherapy-related toxicity. The aim of study was to identify changes in plasma homocysteine (HCy), 5-methyltetrahydrofolate (MTHF), pyridoxal phosphate (PLP), 5-fluorouracil (5-FU), fluorodeoxyuridine (FdUr), and deoxyuridine (dUr) levels, as well as salivary levels of uracil (Ura) and dihydrouracil (UH2) in patients with colorectal cancer (CRC) during adjuvant 5-FU-based chemotherapy. Further, the aim was to evaluate whether these metabolites have the potential to be utilized as risk markers for chemotherapy-related toxicity. PATIENTS AND METHODS: Blood and saliva samples were collected from 24 patients immediately before treatment initiation on days 1 and 2 of cycle 1 (C1-1, C1-2) and cycle 2 (C2-1, C2-2). Additional blood samples were collected 10 and 30 min after 5-FU administration to enable calculation of area under the curve (AUC) values. HCy levels were determined by the chemiluminescent microparticle enzyme immunoassay method, Ura and UH2 levels by HPLC, and MTHF, PLP, 5-FU, FdUr, and dUr levels by LC-MS/MS. RESULTS: Significant shifts in metabolite profiles were observed across both chemotherapy cycles, with most metabolites failing to return to baseline levels between cycles. HCy levels declined progressively, whereas MTHF, PLP, FdUr, and dUr increased. Men had higher HCy levels than women. HCy levels correlated positively with age in women at baseline (r = 0.65, p = 0.029), a relationship not seen in men. HCy also correlated positively with the interval between surgery and initiation of chemotherapy during Cycle 1 (r = 0.60, p = 0.0024 at C1-2). Plasma 5-FU rose at C1-2 in both sexes but diverged during Cycle 2, decreasing in men while continuing to rise in women. Twelve patients required 5-FU dose reduction. These patients had higher plasma HCy and lower salivary Ura at C1-1 and C2-1, along with greater increases in Ura and 5-FU between C2-2 and C1-2 (Δ-values). A predictive model incorporating ΔUra, Δ5-FU, and sex strongly predicted need for 5-FU dose reduction (p < 0.0001). CONCLUSIONS: Significant alterations in metabolite levels occur during the first two cycles of adjuvant 5-FU-based chemotherapy, potentially contributing to treatment-related toxicity. Moderate associations between HCy levels, patient age, and timing of chemotherapy initiation suggest that patient-related factors may influence one-carbon metabolism during 5-FU/LV therapy. Higher Ura and 5-FU levels in women during Cycle 2 may contribute to sex-related differences in treatment-related toxicity. Changes in Ura and 5-FU levels between treatment cycles (ΔUra and Δ5-FU AUC) might be particularly informative for guiding individualized 5-FU dose adjustments, especially when interpreted in the context of patient sex.